Jessica Lebenberg

Nonsupervised Ranking of Different Segmentation Approaches: Application to the Estimation of the Left Ventricular Ejection Fraction From Cardiac Cine MRI Sequences

A statistical methodology is proposed to rank several estimation methods of a relevant clinical parameter when no gold standard is available. Based on a regression without truth method, the proposed approach was applied to rank eight methods without using any a priori information regarding the reliability of each method and its degree of automation. It was only based on a prior concerning the statistical distribution of the parameter of interest in the database. The ranking of the methods relies on figures of merit derived from the regression and computed using a bootstrap process. The methodology was applied to the estimation of the left ventricular ejection fraction derived from cardiac magnetic resonance images segmented using eight approaches with different degrees of automation: three segmentations were entirely manually performed and the others were variously automated. The ranking of methods was consistent with the expected performance of the estimation methods: the most accurate estimates of the ejection fraction were obtained using manual segmentations. The robustness of the ranking was demonstrated when at least three methods were compared. These results suggest that the proposed statistical approach might be helpful to assess the performance of estimation methods on clinical data for which no gold standard is available.

Spatial normalization of brain images and beyond

The deformable atlas paradigm has been at the core of computational anatomy during the last two decades. Spatial normalization is the variant endowing the atlas with a coordinate system used for voxel-based aggregation of images across subjects and studies. This framework has largely contributed to the success of brain mapping. Brain spatial normalization, however, is still ill-posed because of the complexity of the human brain architecture and the lack of architectural landmarks in standard morphological MRI. Multi-atlas strategies have been developed during the last decade to overcome some difficulties in the context of segmentation. A new generation of registration algorithms embedding architectural features inferred for instance from diffusion or functional MRI is on the verge to improve the architectural value of spatial normalization. A better understanding of the architectural meaning of the cortical folding pattern will lead to use some sulci as complementary constraints. Improving the architectural compliance of spatial normalization may impose to relax the diffeomorphic constraint usually underlying atlas warping. A two-level strategy could be designed: in each region, a dictionary of templates of incompatible folding patterns would be collected and matched in a way or another using rare architectural information, while individual subjects would be aligned using diffeomorphisms to the closest template. Manifold learning could help to aggregate subjects according to their morphology. Connectivity-based strategies could emerge as an alternative to deformation-based alignment leading to match the connectomes of the subjects rather than images.

Improved Estimation of Cardiac Function Parameters Using a Combination of Independent Automated Segmentation Results in Cardiovascular Magnetic Resonance Imaging.

This work aimed at combining different segmentation approaches to produce a robust and accurate segmentation result. Three to five segmentation results of the left ventricle were combined using the STAPLE algorithm and the reliability of the resulting segmentation was evaluated in comparison with the result of each individual segmentation method. This comparison was performed using a supervised approach based on a reference method. Then, we used an unsupervised statistical evaluation, the extended Regression Without Truth (eRWT) that ranks different methods according to their accuracy in estimating a specific biomarker in a population. The segmentation accuracy was evaluated by estimating six cardiac function parameters resulting from the left ventricle contour delineation using a public cardiac cine MRI database. Eight different segmentation methods, including three expert delineations and five automated methods, were considered, and sixteen combinations of the automated methods using STAPLE were investigated. The supervised and unsupervised evaluations demonstrated that in most cases, STAPLE results provided better estimates than individual automated segmentation methods. Overall, combining different automated segmentation methods improved the reliability of the segmentation result compared to that obtained using an individual method and could achieve the accuracy of an expert.

An improved FSL-FIRST pipeline for subcortical gray matter segmentation to study abnormal brain anatomy using quantitative susceptibility mapping (QSM)

Accurate and robust segmentation of subcortical gray matter (SGM) nuclei is required in many neuroimaging applications. FMRIBs Integrated Registration and Segmentation Tool (FIRST) is one of the most popular software tools for automated subcortical segmentation based on T1-weighted (T1w) images. In this work, we demonstrate that FIRST tends to produce inaccurate SGM segmentation results in the case of abnormal brain anatomy, such as present in atrophied brains, due to a poor spatial match of the subcortical structures with the training data in the MNI space as well as due to insufficient contrast of SGM structures on T1w images. Consequently, such deviations from the average brain anatomy may introduce analysis bias in clinical studies, which may not always be obvious and potentially remain unidentified. To improve the segmentation of subcortical nuclei, we propose to use FIRST in combination with a special Hybrid image Contrast (HC) and Non-Linear (nl) registration module (HC-nlFIRST), where the hybrid image contrast is derived from T1w images and magnetic susceptibility maps to create subcortical contrast that is similar to that in the Montreal Neurological Institute (MNI) template. In our approach, a nonlinear registration replaces FIRSTs default linear registration, yielding a more accurate alignment of the input data to the MNI template. We evaluated our method on 82 subjects with particularly abnormal brain anatomy, selected from a database of >2000 clinical cases. Qualitative and quantitative analyses revealed that HC-nlFIRST provides improved segmentation compared to the default FIRST method.

The dynamics of cortical folding waves and prematurity-related deviations revealed by spatial and spectral analysis of gyrification

&NA; In the human brain, the appearance of cortical sulci is a complex process that takes place mostly during the second half of pregnancy, with a relatively stable temporal sequence across individuals. Since deviant gyrification patterns have been observed in many neurodevelopmental disorders, mapping cortical development in vivo from the early stages on is an essential step to uncover new markers for diagnosis or prognosis. Recently this has been made possible by MRI combined with post‐processing tools, but the reported results are still fragmented. Here we aimed to characterize the typical folding progression ex utero from the pre‐ to the post‐term period, by considering 58 healthy preterm and full‐term newborns and infants imaged between 27 and 62 weeks of post‐menstrual age. Using a method of spectral analysis of gyrification (SPANGY), we detailed the spatial‐frequency structure of cortical patterns in a quantitative way. The modeling of developmental trajectories revealed three successive waves that might correspond to primary, secondary and tertiary folding. Some deviations were further detected in 10 premature infants without apparent neurological impairment and imaged at term equivalent age, suggesting that our approach is sensitive enough to highlight the subtle impact of preterm birth and extra‐uterine life on folding. Graphical abstract Figure. No caption available. HighlightsThe progression of cortical folding was studied from the pre‐ to post‐term period.SPANGY provided a quantitative spectral and spatial analysis of cortical patterns.We showed three successive folding waves with different characteristic age points.This approach suggested deviations in primary folding in premature infants at TEA.

Exploring the successive waves of cortical folding in the developing brain using MRI and spectral analysis of gyrification

In the developing human brain, gyrification is a complex process going through the successive appearance of primary folds (from 20 weeks of gestational age GA), secondary folds (from 32w GA) and tertiary folds (around term age). While this sequence is finely described in fetuses and preterm newborns of different ages using MRI and folding indices, there is still no fully objective assessment of the folding stage at the individual level. We examined the potential of a new method of spectral analysis of gyrification (SPANGY) that was applied to cortical surfaces of 26 preterm newborns, 9 full-term newborns and 17 infants to quantify the spatial-frequency structure of folding. Based on modelling approaches, we unraveled 4 periods along the developmental sequence from 27 to 62w GA, with relevant timepoints around 31w, 36-38w, and 44-47w GA. These periods showed specific folding features, with spatial patterns of increasing frequencies.

Mapping the asynchrony of cortical maturation in the infant brain: A MRI multi-parametric clustering approach

&NA; While the main neural networks are in place at term birth, intense changes in cortical microstructure occur during early infancy with the development of dendritic arborization, synaptogenesis and fiber myelination. These maturational processes are thought to relate to behavioral acquisitions and the development of cognitive abilities. Nevertheless, in vivo investigations of such relationships are still lacking in healthy infants. To bridge this gap, we aimed to study the cortical maturation using non‐invasive Magnetic Resonance Imaging, over a largely unexplored period (1–5 post‐natal months). In a first univariate step, we focused on different quantitative parameters: longitudinal relaxation time (T1), transverse relaxation time (T2), and axial diffusivity from diffusion tensor imaging (&lgr;//) These individual maps, acquired with echo‐planar imaging to limit the acquisition time, showed spatial distortions that were first corrected to reliably match the thin cortical ribbon identified on high‐resolution T2‐weighted images. Averaged maps were also computed over the infants group to summarize the parameter characteristics during early infancy. In a second step, we considered a multi‐parametric approach that leverages parameters complementarity, avoids reliance on pre‐defined regions of interest, and does not require spatial constraints. Our clustering strategy allowed us to group cortical voxels over all infants in 5 clusters with distinct microstructural T1 and &lgr;// properties The cluster maps over individual cortical surfaces and over the group were in sound agreement with benchmark post mortem studies of sub‐cortical white matter myelination, showing a progressive maturation of 1) primary sensori‐motor areas, 2) adjacent unimodal associative cortices, and 3) higher‐order associative regions. This study thus opens a consistent approach to study cortical maturation in vivo. Graphical abstract Figure. No caption available. HighlightsThe cortical maturation was studied in infants between 1 and 5 months of age.Diffusion and relaxometry MRI were analyzed in univariate and multivariate manner.Clustering analyses provided reliable results both at the subject and group levels.Distinct maturation profiles were shown across regions of the infant cortex.

Comparison of different segmentation approaches without using gold standard. Application to the estimation of the left ventricle ejection fraction from cardiac cine MRI sequences

A statistical method is proposed to compare several estimates of a relevant clinical parameter when no gold standard is available. The method is illustrated by considering the left ventricle ejection fraction derived from cardiac magnetic resonance images and computed using seven approaches with different degrees of automation. The proposed method did not use any a priori regarding with the reliability of each method and its degree of automation. The results showed that the most accurate estimates of the ejection fraction were obtained using manual segmentations, followed by the semiautomatic methods, while the methods with the least user input yielded the least accurate ejection fraction estimates. These results were consistent with the expected performance of the estimation methods, suggesting that the proposed statistical approach might be helpful to assess the performance of estimation methods on clinical data for which no gold standard is available.

A framework based on sulcal constraints to align preterm, infant and adult human brain images acquired in vivo and post mortem

Robust spatial alignment of post mortem data and in vivo MRI acquisitions from different ages, especially from the early developmental stages, into standard spaces is still a bottleneck hampering easy comparison with the mainstream neuroimaging results. In this paper, we test a landmark-based spatial normalization strategy as a framework for the seamless integration of any macroscopic dataset in the context of the Human Brain Project (HBP). This strategy stems from an approach called DISCO embedding sulcal constraints in a registration framework used to initialize DARTEL, the widely used spatial normalization approach proposed in the SPM software. We show that this strategy is efficient with a heterogeneous dataset including challenging data as preterm newborns, infants, post mortem histological data and a synthetic atlas computed from averaging the ICBM database, as well as more commonly studied data acquired in vivo in adults. We then describe some perspectives for a research program aiming at improving folding pattern matching for atlas inference in the context of the future HBP’s portal.